📊 DATA602: Principles of Data Science – Fall 2024

This repository contains a collection of Jupyter notebooks created as part of the DATA602 course at the University of Maryland. The course emphasizes real-world applications of data science techniques such as data cleaning, visualization, regression, classification, clustering, and time series forecasting using Python.

---

📁 Contents

1. Assignment 3 – Intro & Preprocessing

Exploratory Data Analysis and basic data preparation tasks.

2. Assignment 4 – Feature Engineering

Feature extraction and transformation of raw input data into structured format.

3. Assignment 5 – Bias and Resampling

Focuses on identifying algorithmic bias and using resampling techniques.

4. Assignment 6 – Regression Models

Implementation of linear, logistic, and LASSO regression using scikit-learn.

5. Assignment 7 – Classification

Binary classification tasks including threshold optimization and confusion matrix evaluation.

6. Clustering Homework

K-means and k-NN clustering and classification.

7. SVM Homework

Application of Support Vector Machines with kernel tuning.

8. Time Series Homework

Forecasting models using ARIMA, STL decomposition, and seasonal trends.

9. Data602_1 – Initial Explorations

Early EDA and initial experimentation.

---

🛠 Tools & Libraries

• Python, Jupyter Notebooks
• Pandas, NumPy, Matplotlib, Seaborn
• scikit-learn, XGBoost
• Statsmodels (for time series)

---

📂 Structure

DATA602-Principles-of-Data-Science